Recently, an anonymous borrower who goes by the screenname "compassion-engine220" sought out a loan from Prosper.com. 

From the documentation, this borrower isn't exactly the best risk. With revolving debt of $10,429 -- or 79% of his or her bankcard limits -- and two delinquencies amounting to $875, this borrower won't get a loan at a prime interest rate. Not helping here is the borrower's income, reported to be $25,000 to $49,999 per year, earned from a nine-year career as a nurse's aide.

Despite some credit issues, Prosper was able to fund an $18,000 loan at an interest rate of 25% per year, to be repaid in 60 monthly installments of $502.07.

Whether a borrower will repay a loan depends on two factors: their ability and willingness to repay. We can't judge compassion-engine220's willingness to repay from a few data points. But we do know their ability to repay hinges on the borrower's reported job and income.

Is compassion-engine220 really a nurse's aide who earns $25,000 to $49,999 per year? 

The truth is that no one really knows.

No proof, plenty of capital
Increasingly, peer-to-peer lenders are lending billions of dollars to marginal borrowers with little verification that the borrowers are who they purport to be. But that hasn't stopped peer-to-peer marketplaces from finding investors willing to buy packages of loans they originate.

Prosper's S-1 filed with the SEC reveals that it verified employment and/or income for only 59% of the loans originated through its marketplace during a period spanning from 2009 to 2015. Of these loans, Prosper notes that it cancelled 15% of loan listings, or roughly one out of every six loans where it sought additional information. (Hat tip to FatTailCapital for bringing this issue to light.)

Prosper isn't the only company that does limited due diligence on its borrowers. LendingClub (LC -0.12%) noted in its most recent annual report that it seeks to verify the applicant's identity, income, or employment with "various data sources" or "by contacting the human resources department" of the borrower's employer.

However, the process seemingly stops there. Deeper in its report, LendingClub notes that it "often does not verify a borrower's stated tenure, job title, home ownership status, or intention for the use of loan proceeds."

A boom with no limits
To be fair, small-dollar loans typically lack the underwriting controls of larger loans, such as mortgages. Credit card applications are frequently approved instantly over the Internet, without any further information beyond what is provided on the application. As a result, losses in credit card portfolios tend to peak shortly after a marketing spree for new customers, as bad risks immediately surface. "Good" borrowers receive credit limit increases, while "bad" borrowers get cut off.

But where credit card balances have largely stagnated and the bank that loans the money is responsible for reviewing the documentation, there's a modern gold rush in peer-to-peer loans originated by marketplaces with very little skin in the game.

The CEO of LendingTree (TREE 0.27%) once said on a conference call that personal loan lenders have "7 to 10 times more money to lend than they can actually lend." Historically, when the supply of loanable funds exceeds demand, originators create demand, often by lowering their credit standards.

Peer-to-peer lenders aren't the only lenders partying like it's 2005. Small-business lender On Deck Capital (ONDK) has originated billions of dollars of small-business loans at high interest rates. In the first nine months of 2015, more than $194 million of its originations came from unpaid principal balances rolled into new loans. Its customers rolled $106 million of balances into new loans in the prior-year period.

Given steep origination fees and high interest rates, one has to wonder if On Deck's customers are rolling over balances only because it is their only option to keep current. What might its loan losses look like if it didn't roll over balances for its customers?

The banking industry is boring, and at times, archaic, seemingly ripe for disruption. But behind the bank branches are hundreds of years of collective underwriting experience. And while the big, bad banks certainly get things wrong from time to time, they have what the online lenders do not: Years and years of experience, and a collective history that warns fast loan growth often foreshadows large loan losses.

Luckily, online lending is still a drop in the bucket. The industry's typical customer borrows a few thousand dollars -- relative pocket change, even in the world of consumer finance. An implosion in online lending wouldn't have the cataclysmic effect that the mortgage bust had on the American economy.

But those who have a stake in online lending -- either by holding shares of top marketplaces or by directly investing in loans through their online accounts -- should take a step back. Ask yourself if maybe, just maybe, the hottest arena in finance has become a little overheated.